Fully Decentralized Approximate Dynamic Programming for Stochastic Energy Management of a Networked Microgrid System

With the increasing penetration of renewable generation into power system, the energy management of multi-microgrid (MMG) system has become an important problem. In this paper, a bi-level distributionally robust energy management model for MMG system with uncertain renewable generation is proposed. The distribution system operator (DSO) and each microgrid (MG) are considered as different entities with various objectives. In each MG, the Wasserstein distance based ambiguity set is utilized to describe the uncertain distribution of renewable generation. Due to the coupling power exchange, the alternating direction method of multipliers (ADMM) is applied to solve the problem in a decentralized manner. For individual MG energy management, the column and constraint generation (CCG) algorithm is used to solve it. Simulation experiments are carried out to validate the effectiveness of the proposed model based on a modified IEEE 33-bus system with three MGs.

Distributed robust operational optimization of networked microgrids embedded interconnected energy hubs

Energy management of cooperative microgrids: A distributed optimization approach

The Cox proportional hazards model is a popular semi-parametric model for survival analysis. In this paper, we aim at developing a federated algorithm for the Cox proportional hazards model over vertically partitioned data (i.e., data from the same patient are stored at different institutions). We propose a novel algorithm, namely VERTICOX, to obtain the global model parameters in a distributed fashion based on the Alternating Direction Method of Multipliers (ADMM) framework. The proposed model computes intermediary statistics and exchanges them to calculate the global model without collecting individual patient-level data. We demonstrate that our algorithm achieves equivalent accuracy for the estimation of model parameters and statistics to that of its centralized realization. The proposed algorithm converges linearly under the ADMM framework. Its computational complexity and communication costs are polynomially and linearly associated with the number of subjects, respectively. Experimental results show that VERTICOX can achieve accurate model parameter estimation to support federated survival analysis over vertically distributed data by saving bandwidth and avoiding exchange of information about individual patients. The source code for VERTICOX is available at: https://github.com/daiwenrui/VERTICOX.

Developing Hybrid Demand Response Technique for Energy Management in Microgrid Based on Pelican Optimization Algorithm

With the increasing interdependence of various energy carriers, the operation of power systems is found to correlate closely with the limitations on the other energy infrastructures. This paper presents a mixed-integer linear programming (MILP) model for the microgrid (MG) optimal scheduling considering technical and economic ties between electricity and natural gas (NG) systems. In the proposed methodology, different energy converters and storages, including combined heat and power (CHP) units, electricity/heat storage units, and distributed energy resources (DERs) are considered. The proposed model allows the MG operator to minimize the operation cost of the MG while different operational limitations on the energy hub are satisfied. The model is developed based on AC power flow constraints so as to respect reactive power and voltage security constraints. The efficiency and robustness of the proposed MILP formulation are successfully verified using a large-scale test MG.

Networked microgrids (NMGs) are developing as a viable approach for integrating an expanding number of distributed energy resources (DERs) while improving energy system performance. NMGs, as compared to typical power systems, are constructed of many linked microgrids that can function independently or as part of a more extensive network. This allows NMGs to be more flexible, dependable, and efficient. The present study comprehensively investigates architecture, communication, and cybersecurity issues in NMGs. This comprehensive study examines various aspects related to networked microgrids (NMGs). It explores the architecture of NMGs, including control techniques, protection, standards, and the challenges associated with their adoption. Additionally, it investigates communication in NMGs, focusing on the technologies, protocols, and the impact of communication on the functioning of these systems. Furthermore, this study addresses cybersecurity challenges specific to NMGs, such as diverse cyberattack types, detection and mitigation strategies, and the importance of awareness training. The findings of this study offer valuable insights for NMG researchers and practitioners, emphasizing critical aspects that must be considered to ensure the safe and efficient operation of these systems.

This paper addresses the optimal energy management and scheduling of renewable based microgrids (MGs) in the highly correlated uncertain environments. The proposed MG incorporates different types of dispatchable and non-dispatchable units such as micro turbine (MT), fuel cell (FC), wind turbine (WT), photovoltaic panel (PV) and a battery bank (BAT). Due to the nonlinear and non-convex nature of the proposed problem, a heuristic approach based on the bird mating optimization (BMO) algorithm is employed to solve the problem. The uncertainties associated with the market price (MP), PV output power, WT output power and hourly load uncertainties are modeled using the unscented transformation (UT) method. In addition, two sampling procedures are adopted, namely as the (2m+1) and the (2m) point methods, to provide a comparative analysis showing the highly capabilities of the proposed framework. Moreover, the correlation among uncertain variables is taken into consideration and their effects on the MG's optimal operation are analyzed. The simulation results reveal the high accuracy, effectiveness, relatively timesaving and tending to avoid local minimums by the proposed model.

The network identification plays a very prominent role for the network operator to accomplish the various objectives such as state-estimation, monitoring, control, planning, and real-time analytics. The network structure varies from time-to-time and its details are often not available with the network operator. To address this issue, an alternating direction method of multipliers (ADMM) based framework is presented herein to identify the network topology and line parameters using smart meter (SM) and micro phasor measurement unit (micro-PMU) measurements. The presented algorithm is divided into two sections, 1) approximate parameter evaluation through regression, to extract the partial topology information, and 2) complete network topology identification through the ADMM framework. This algorithm accomplishes the objectives of identifying the network configuration, branch parameters (e.g. conductance and susceptance), and change in branch parameters. Simulation results demonstrate the effectiveness of the presented algorithm on the benchmarked IEEE 13-bus and IEEE 123-bus feeders under various operating scenarios. Furthermore, the presented framework illustrates excellent network identification even with the presence of the stochastic nature of renewable power generation. The presented algorithm exhibits an excellent performance even with the consideration of noise in both measurements. In addition, the comparative performance is carried out on the benchmarked unbalanced IEEE 13-bus and balanced IEEE 33-bus feeders to highlight the efficacy of the presented framework over the state-of-art framework.

Poisson image deblurring with frame-based nonconvex regularization

A low-complexity ADMM-based secure beamforming for IRS-assisted uplink cognitive radio networks with NOMA

Synchronous and asynchronous parallel computation for large-scale optimal control of connected vehicles

SummaryExtreme learning machine (ELM) is a prominent example of neural network with its fast training speed, and good prediction performance. An online version of ELM called online sequential extreme learning machine (OS‐ELM) has also been proposed for the sequential training. Combined with the need for regularization to prevent over‐fitting in addition to the large number of neurons required in the hidden layer, OS‐ELM demands huge amount of computation power for the large‐scale data. In this article, a mixed norm () regularized online machine learning algorithm (MRO‐ELM) that is based on alternating direction method of multipliers (ADMM) is proposed. A linear combination of the mixed norm and the Frobenius norm regularization is applied using the ADMM framework and update formulas are derived. Graphics processing unit (GPU) accelerated version of MRO‐ELM (GPU‐MRO‐ELM) is also proposed to reduce the training time by processing appropriate parts in parallel using the implemented custom kernels. In addition, a novel automatic hyper‐parameter tuning method is incorporated to GPU‐MRO‐ELM using progressive validation with GPU acceleration. The experimental results show that the MRO‐ELM algorithm and its GPU version outperform OS‐ELM in terms of training speed, and testing accuracy. Also, compared to the cross validation, the proposed automatic hyper‐parameter tuning demonstrates dramatical reduction in the tuning time.

GCN-based proximal unrolling matrix completion for piecewise smooth signal recovery

Grid-interactive efficient buildings can provide many benefits for the power grid. However, it is very challenging to coordinate their operations intelligently due to many considerations, e.g., privacy protection and scalability. To this end, this paper investigates an optimal coordination operation problem of multiple grid-interactive residential buildings under uncertainties to minimize the total peak demand violation of all residential buildings during the grid service periods while minimizing their respective operation cost. Taking into account multi-source uncertainties, temporal and spatial couplings, inexplicit thermal dynamics models, and the lack of information sharing among residential buildings incurred by privacy protection, it is difficult to design a centralized model-based optimization operation method. To overcome the challenge, we design an online distributed coordination operation algorithm based on the alternating direction method of multipliers (ADMM) and relative entropy regularized policy iteration (RERPI). The designed algorithm consists of two stages, i.e., independent offline training of each RERPI-based agent related to the building energy management system controller (BEMSC) in each residential building (stage-1) and online distributed execution for coordination between the aggregator controller and each BEMSC under the ADMM framework (stage-2). Simulation results verify the effectiveness and scalability of the designed algorithm.